Statistical Theory
(Selected Papers)
Objective Bayes with Pseudo-Likelihoods
Matching priors are easy to derive from Pseudo-Likellihoods, which eliminates nuisance parameters.
- L. Ventura, S. Cabras, W. Racugno. Default prior distributions from quasi- and quasi-pro le likelihoods. JSPI, 140(11):2937 - 2942, 2010
- S. Cabras, M. E. Castellanos,
W. Racugno, L. Ventura. A matching prior for the shape
parameter of the skew-normal distribution. Scandinavian Journal of Statistics,
39(2):236 - 247, 2012
The goodness of Fit (GOF)
Calbritated p-values are needed to have GOF-reliable measures
- S. Cabras, M. E. Castellanos, O. Ratmann. Goodness of fit for models with intractable likelihood. Test, (in press), 2021
- S. Cabras, M. Castellanos. P-value calibration in multiple hypotheses testing. Statistics in Medicine, 36(18), 2875-2886, 2017
- S. Cabras, M.E. Castellanos, A. Quiros. Goodness-of- t of conditional
regression models for multiple imputation. BA, 6(3):429 - 456,
2011
- S. Cabras, M. Castellanos. Default bayesian goodness-of- t tests for the skew-normal model. JAS, 36(2):223 - 232, 2009
Multiple Testing and Model Selection
Joint evidence from a multiplicity of tests can be obtained by combining modern Bayesian model selection techniques
Multiple Testing:
- S. Cabras and M. E. Castellanos. Transfer Learning in Multiple Hypothesis Testing. Entropy 26, 1, 2024
- S. Cabras. A markov chain representation of the multiple testing problem. SMMR, 27(2):364-383, 2018
- S. Cabras. A note on multiple testing for composite null hypotheses. JSPI, 140:659 - 666, 2010
- F. Bertolino, S. Cabras, M. E. Castellanos, W. Racugno. Unscaled bayes factors for multiple hypothesis testing in microarray experiments. SMMR, 24(6):1030 - 1043, 2015
Model Selection:
- G. García-Donato, S. Cabras, M.E. Castellanos. Model uncertainty quantification in Cox regression. Biometrics, 2023
- M.E. Castellanos, G. García-Donato, S. Cabras. Model Selection Approach for Variable Selection with Censored Data. Bayesian Analysis 16, 271-300, 2021
- S. Cabras. A Dirichlet Process Prior Approach for Covariate Selection. Entropy 22(9), 948, 2020
- C. Armero,S. Cabras, M. E. Castellanos, A. Quirós. Two-stage Bayesian approach for GWAS with known genealogy. JCGS, 28(1): 197-204, 2019
- S. Cabras, W. Racugno, L. Ventura. Higher order asymptotic computation of bayesian significance tests for precise null hypotheses in the presence of nuisance parameters. JSCS, 85(15):2989 - 3001, 2015
- S. Cabras, M.E. Castellanos, S. Perra. A new minimal training sample scheme for intrinsic bayes factors in censored data. CSDA, 81:52 - 63, 2015
- S. Cabras, M.E. Castellanos, S. Perra. Comparison of objective bayes factors for
variable selection in parametric regression models for survival analysis. SIM, 33(26):4637 - 4654, 2014
- S. Cabras, G. Mostallino, W. Racugno. A non-parametric bootstrap
test for the equality of coe cients of variation. Communications in Statistics, 35(3):715 - 727, 2006
Bayesian Modeling
Bayes setting is useful for model building.
Spatial Statistics:
Deep Learning and Bayes (Covid-19):
Extremes:
- M.E. Castellanos, S. Cabras. A default bayesian procedure for the generalized pareto distribution. JSPI, 137(2):473 - 483, 2007
- S. Cabras, J. Morales. Extreme
value analysis within a parametric outlier detection framework. AMSBI, 23(2):157 - 164, 2007
- S. Cabras, M.E. Castellanos, D. Gamerman. A default bayesian approach for regression on extremes. Statistical Modelling, 11(6):557 - 580, 2011
- S. Cabras, M.E. Castellanos. A bayesian approach for estimating extreme quantiles under a semiparametric mixture model. ASTIN Bulletin, 41(1):87 - 106, 2011
Survival:
ABC Modeling
Approximate Bayesian Computation: optimal sampling and GOF
- S. Cabras, M. E. Castellanos, E. Ruli. Approximate bayesian com-putation by modelling
summary statistics in a quasi-likelihood framework. BA,
10(2):411 - 439, 2015
- S. Cabras, M. E. Castellanos, E. Ruli. A quasi likelihood
approximation of posterior distributions for likelihood-intractable complex
models. Metron, 72(2):153 - 167, 2014